INDEX
Explanations
instances of education-related events or accomplishments, particularly graduation
instances of the word "graduated" or related terms indicating completion of academic milestones
New Auto-Interp
Negative Logits
erness
-0.70
hed
-0.67
BOOK
-0.67
bro
-0.66
pro
-0.65
sem
-0.64
drawn
-0.62
pers
-0.61
Reprodu
-0.60
core
-0.60
POSITIVE LOGITS
uates
1.01
uating
0.86
ploma
0.86
graduation
0.81
inois
0.79
icient
0.77
SCHOOL
0.73
Pradesh
0.72
graduate
0.71
avorite
0.71
Activations Density 0.015%